Chapter 6

WATER WAVE EQUIVALENT OF MACH-REFLECTION

R. L. Wiegel
Professor of Civil Engineering
University of California
Berkeley, California

SUMMARY

Periodic (shallow water and transitional water) and solitary water

gravity waves do not reflect from a wall in the manner commonly supposed,
when the angle between the direction of wave advance and the wall is less
than about 5? to 45 degrees. The wave front bends near the wall, becom-
ing normal to the wall, with a small reflected wave. For angles less
than about 20 degrees the reflected wave becomes almost negligible. The
portion of the wave near the wall (called the Mach-stem in air blast
waves) increases in height as the wave continues to move along the wall.
Once the Mach-stem is formed, it will continue to grow even when the wall
is bent around through almost 90 degrees; for periodic waves a Mach-
reflected wave also develops. The Mach-stem is insensitive to undulation,
of the wall. Results of studies of this phenomenon in the laboratory are
presented, together with some observation of its occurrence in the ocean.
The importance of this phenomenon to the study of tsunami action at Hilo,
Hawaii, is presented.

INTRODUCTION

On occasions, tsunamis in certain coastal regions have exhibited

characteristics which do not seem to be accounted for when they are
studied by means of the commonly used types of refraction drawings, dif-
fraction calculations and run-up theory and measurements. It was sug-
gested to the author by Professor John D. Isaacs in 1956 that this might
be due to something analogous to the Mach-reflection phenomenon in acous-
tics. A number of aspects of this phenomenon have been investigated
experimentally by several graduate students under the author's direction
(Perroud, 1957; Chen, 19ol; Slgurdsson and Wiegel, 1962; Nielsen, 1962).
The first tests were made using a solitary wave, as it was believed that
this was most nearly the water wave equivalent of a shock wave in com-
pressible flow.

The first experiments were performed! with a solitary wave incident

to an oblique, vertical, impervious, smooth barrier; it was found that
the phenomenon did exist. Later, tests were performed with periodic
waves, and the Mach-reflection also occurred for these for values of
L/d > about 2 (where L is the wave length and d is the water depth).
Several experiments were performed with types of models which were of
importance to coastal engineering problems. One of these was an undu-
lating vertical wall, which was found to affect only the small details of

82
 WATER WAVE EQUIVALENT OF MACH-REFLECTION 83

the wave motion. A second model consisted of a curved, vertical imper-

vious barrier (similar to curved vertical wall jetties or breakwaters).
Experiments in the laboratory showed that once the Mach-stem formed, it
became so strong that it ultimately became independent of the incident
wave.

Finally, experiments were made with an undistorted model of the bay

at Hilo, Hawaii, and a phenomenon was found to exist which had the appear-
ance of a Mach-reflection.

The only theory available on the Mach-reflection, to the author's

knowledge, is due to Lighthill (l9i+9)> and it is not useful in its pres-
ent form in the solution of water gravity wave problems.

MACH-REFLECTION, SOLITARY tfAVE

It is usually assumed that a wave which encounters an obstacle will

either reflect from it, be dissipated, or both. In certain cases, how-
ever, a third possibility exists: the crest of the wave near the wall
may bend, becoming normal to it, with the energy density adjacent to the
wall increasing as the wave moves along. This phenomenon is known as a
Mach-reflection.

Perroud (1957) made a series of tests in the laboratory of the

reflection of a solitary wave from a straight, vertical, impervious wall.
He found that three types of patterns occurred, with one of the patterns
being a special case of one of the others. The critical angle of inci-
dence, i, separating these two types of patterns appeared to be 45 degrees
(Chen, 196l, found the critical angle to be between 55 and 40 degrees).

For incident angles (the angle between the direction of wave advance
and the wall) greater than 35 to 45 degrees the reflection pattern is
"normal" (Fig. l). The incident and reflected waves are slightly dis-
turbed near the wall, but the angle of reflection is equal to the angle
of incidence, and the reflected wave height is only slightly less than
the incident wave height. However, the reflected wave is followed by a
trough, except for the case in which the angle of incidence is 90 degrees.
It is interesting to note that in the latter case the wave height at the
wall was 20 percent higher than twice the incident wave height.

For angles of incidence less than about 55 to 45 degrees the reflec-

tion appears to be of the type called a Mach-reflection in acoustics, v/hen
the angles of incidence are less than about 20 degrees, the wave crest
bends, becoming normal to the wall, and no reflected wave appears, 'v/hen
the angle of incidence is greater than 20 degrees but less than 40 to 45
degrees, three waves are present, the incident wave I, a Mach-reflected
wave R, and the Mach-stem wave M, the width of which grows as the wave
moves along the wall. The angle 6 (5 = r - i) depends upon the angle of
incidence in the manner shown in Fig. 2. The reflected wave height is
smaller than the incident wave height, and the angle of reflection (r) is
greater than the angle of incidence. The height of the Mach-stem portion
is greater than the incident wave height and is at its maximum at the wall.
The Mach-reflected and Mach-stem waves are followed by a trough.
84 COASTAL ENGINEERING

(A) i < 20° (B) 20°<i<45° (C) i>45°

Mach-stem without Mach-stem with Regular reflection
reflection reflection

PIG 1. REFLECTION OP A SOLITARY WAVE WITH VERTICAL BREAKWATER.

I : Incident wave
R : Reflected wave

o 1

>
\
\
N
\
\
\
1

i—— — < i

15' 30° <45" 60' 75* 90° '0 7i 15° 22i"

Angle ftf incidence i
PIG 2Q. STEM-ANGLE i POR SOLITARY WAVE. PIG 2b. HEIGHT OP STEM-
EACH PLOT IS AN AVERAGE VALUE POR 6 WAVE POR SOLITARY WAVE.
DIFFERENT WAVE HEIGHTS, (%/d: .05-.43). Water depth d - 0.132 ft
Water depth d • 0.132 ft. ( From Perroud, 1957)
 WATER WAVE EQUIVALENT OF MACH-RE FLECTION 85
In practice, most structures and coastal areas have slopes. Experi-
ments made with smooth impermeable slopes (Chen, 196l) showed that for
nearly vertical walls the phenomenon looked the same as for the case of
a vertical wall (3 = 90 degrees). As the wall slope (3) was decreased
a large horizontal eddy formed over the slope, and as the slope was
decreased further, the wave broke along the slope (Fig. j); the slope
angle at which this occurred was found to he dependent upon the angle of
incidence. The values of i and 3 which determine whether or not the wave
breaks over the slope are shown in Fig. k, for a specific value of H/d.
For small values of H/d (where H is the incident wave height and d is the
water depth) no break occurred when 0 ^ 90 degrees. For intermediate
values of H/d it was not possible to determine whether or not a wave
would break because of the small size of the tank. In regard to this,
Friedlander (1946) has shown theoretically for a sound pulse incident to
a wedge (less than 90 degree wedge) that the pulse has to travel a con-
siderable distance along the wedge before the pressure at the wedge
builds up to its maximum value of twice the pressure of the incidence
pulse, and that the smaller the wedge angle (hence, the smaller i) the
greater this distance will be. For example, for a 30-degree wedge the
pulse must travel nearly 60 pulse lengths to build up to about 90 percent
of its final value.

Waves incident to an overhanging wall (3 > 90 degrees) behaved in a

manner similar to a vertical wall except that the stem did not grow with
distance from the start as was found by Perroud for the vertical wall.
At least, the width of the Mach-stem did not appear to grow within the
limits of the experimental facilities.

The region of i and 3 for which various types of reflections occur

are shown in Fig. 5.

The author (Wiegel, 1963) has observed ocean waves with the appear-
ance of a Mach-stem occurring along a curved structure, and this led to
another series of laboratory studies. A vertical wall breakwater was
made of a piece of sheet metal. A 2 ft long straight section, starting
at one wall of the tank, was placed at a 12 degree angle with the wall
(Sigurdsson and Wiegel, 1962). This straight section was connected
tangentially to a 66 degree segment of a circle of 3-33 ft radius, to
make a total change in direction of 16 degrees (Fig. 6). Tests made with
a barrier with a smooth wavy surface (corrugated aluminum, with corruga-
tions 0.021 ft deep by 0.135 ft long, about l/2 and 3 times the incident
wave height, respectively) and with a rough barrier showed similar results
(Fig. 7)- It was observed that the amplitudes at the barrier were greater
in the hollows than in the ridges of the corrugations.

The essential feature that was determined from the series of tests
just described is that once the Mach-reflection starts, with an angle of
incidence less than the required value, the non-linear Mach-stem becomes
so strong that it continues to move around the curved barrier, normal to
the barrier, and near the end of the barrier becomes independent of the
incident wave. In this region there were essentially two separate waves,
the "incident wave" and the Mach-stem advancing at an appreciable angle
(up to 90 degrees) to the "incident wave." These two waves were connected
86 COASTAL ENGINEERING

SKETCH OF BREAK WITH EDDY

FIGURE 3a

FIGURE 3 b-WAVE BREAK

Wave Break
(near critical angle
at which break ceases)

SKETCH OF BREAK WITHOUT EDDY (After Chen, 1961)

FIGURE 3c
FIGURE 3 Q.b.c
 WATER WAVE EQUIVALENT OF MACH-RE FLECTION 87

70
a = 0 16'
HT
60 -f
a = 0 325
(pos 8)
50
Non- Breaking Region

40
_0__ D
-c—
T3
•30 OVv,

20
E reakin g Reg on
N i\

10

0 10 20 30 40 50 60 70 80 90 100 I
P, degrees
(after CHEN, 1961)

FIGURE 4 ANGLE OF INCIDENCE SEPARATING BREAKING

AND NON-BREAKING REGIONS

60
d = 0.16'
50 d
0.275

R egulor Reflec ed Wo* e

40
(A
a>
a>
y K loch R eflectio n

o> 30 ^
a> "ft—1i~*v~'
•o
t> lo Ref ection
•- 20
No Re fleeted Wove

10

50 60 70 80 90 100 110 120 130 140 150 160

p , degrees
(after CHEN, I960

FIGURE 5 INCIDENT ANGLE I BEYOND WHICH REGULAR

REFLECTED WAVE OCCURS FOR DIFFERENT SLOPES 0,
AND i AT WHICH MACH REFLECTION STARTS FOR
, EACH NEGATIVE SLOPE ( fi > 90° ).
88 COASTAL ENGINEERING

3.0

CREST POSITIONS

1.0 2.0 3.0 4.0

DISTANCE ALONG CREST (S)-FEET

FIGURE 6. CREST POSITION AND CREST ELEVATION

SERIES E
(From Sigurdsson and Wiegel, 1962)
 WATER WAVE EQUIVALENT OF MACH-REFLECTION 89

a •^ •0<
1 1 1 to
U3IMyv8 JO 0N3 0>

^
/f
<( a i
7
30 m
o
in
c
O
to
V>
•o
CO ^
a: a>
UJ (7)
or
or b
o
<
m 3
_i
<
O o
en <fr •- H
UJ ce
UJ UJ
>
< 9 1
> or
or
Ul Ul
UJ z
or o
X X
i- i-
or 1-
z < <
§ >
Om UJ
to _l
— O t\j in in <* O
00 CO — 00 oo en Ul
M * — <V CM CM Z
o UJ
0 d o 6 o o -1 >
<
•I ii ii ii H M
I
5^^^ Ul
o s
X X X l" X' X z
<
1-
2
or X
UJ <
2
GE
<
CO

1
-r-

1 " " *I ' «o

to

-0-4A
< < 4 Ul UJ U.
to to to to to to
UJ UJ W Ul UJ UJ
or or or or or a:
UJ Ul Ul Ul Ul u
to to to to w to
o • e o • 0

o CO to 00 to CO to CVI O
CO CM C\j d o d o
('H/H) 1H9I3H 3AVM 1N30I0NI / 1H9I3H 3AVM
FIGURE 7
90 COASTAL ENGINEERING

by a transition zone, and in this zone the incident and stem waves were
superimposed, and there was a considerable variation in the profile.

When a barrier was made of 3/8 inch gravel to the same plan as the
impervious barrier, but with a 36 degree side slope, there was no indica-
tion of either a build up of wave amplitude at the barrier, or of a
reflected wave.

MACH-REFLECTION, PERIODIC WAVES

The experiments described so far were for solitary waves, as the

solitary wave is analogous to the single pressure impulse in acoustic
waves in air. The phenomenon had been observed for periodic waves in
some coastal areas, so it was known that it did exist in actual condi-
tions for nearly periodic waves. In order to see whether or not it
existed for deep water as well as for shallow water waves, a program of
laboratory studies was undertaken (Nielsen, 1962).

The first series of tests were made in the same tank as the one used
in the various tests described in the previous section. Most of the tests
were made using the same water depth, 0.160 ft, but some tests were made
in deeper water, 0.2^0 ft. It was found that for shallow water waves a
Mach-stem formed which was similar to the one formed by the solitary wave.
This was true even for waves in the transitional region, as can be seen
in Fig. 8a in which the phenomenon is shown for an L/d of 4.9 for angles
of incidence of 5> 10 > 15> and- 20 degrees. In these shadow photographs
the dark lines are the wave crest, and the wider the "line" the higher
the wave. In deep water a type of Mach-stem forms, but it occurs to the
rear of the incident wave and is connected to the incident wave by a
crest with a compound curve, as can be seen in Fig. 8b. In addition, a
curved Mach-reflected type of wave also occurs which is similar in appear-
ance to the Mach-reflected wave of a solitary wave when 20 degrees
< i < 35 to k3 degrees. It is different in that it locks in with the
intersection of the Mach-stem and incident wave of the following wave.
For deeper water waves, L/d < 2, it appears that the reflected wave
gradually develops into something similar to a normal type of reflection
pattern, except that there appears to be a small Mach-stem adjacent to
the barrier. For example, as can be seen in Fig. 8c for i > 5 to 10
degrees, the more or less regular reflection pattern seems to be emerg-
ing, while for i = 5 degrees only the Mach-stem occurs.

It was found that the width of the Mach-stem increased as the wave
moved along the barrier, as was the case for the solitary wave. It was
found that the longer the wave length the greater the width for a given
distance along the barrier; however, it was found that the solitary wave
had the minimum width for a given distance, which is quite surprising as
one normally thinks of a solitary wave having a long "length."

The ratio of wave height of the stem at the barrier to the incident
wave height, H/HJ, was a complicated function of distance along the bar-
rier, water depth and wave length. In general, for a given angle of
incidence it increased up to a certain distance along the barrier, and
92 COASTAL ENGINEERING

then remained nearly constant, and in some instances started to decrease.

The effect of the angle of incidence upon H/HT was similar to the case
for the solitary wave, as can be seen in Fig. 9-

In order to study the phenomenon near the end of a curved break-

water, it was necessary to use a larger facility so that there would be
no reflections from the wall opposite the barrier until a large number
of waves could be measured. These experiments were made in a 150 ft long
by 6k ft wide by 2 1/2 ft deep model basin. A vertical impervious bar-
rier was constructed similar to the one described in a previous section,
but about five times as large, as can be seen in Fig. 10. A photograph
of the waves that were observed is given in Fig. 11.

The results were similar to the results obtained through the use of
the small model basin. Color motion pictures were taken (l6 mm), and
they were studied thoroughly. Use of these motion pictures, together
with the other types of results, permitted Nielsen to determine the
details of the motion of a system of periodic waves moving along a curved
barrier. His conclusions, in graphical form, are given in Fig. 12. The
Mach-stem forms on the straight portion of the barrier (A), dropping
behind the wave crest (which is different from the case of the solitary
wave). As it moves along the curved section, a Mach-reflected wave also
forms (B). AS the wave moves still farther along the curve, the Mach-
stem and Mach-reflected waves seem to coalesce (C), and then separate
from the barrier as a free wave (D) moving nearly normally to the direc-
tion of the incident wave. At this stage the free wave is independent
of the incident wave. The effect of this transition from one type of
wave to another on the wave height along the barrier can be seen in
Fig. 13. The amplitude increases in height gradually as it moves along
the barrier, and then increases rather rapidly as the Mach-stem and Mach-
reflected waves coalesce. Then, as the wave "separates" from the barrier
the height at the barrier decreases rapidly, although the height a short
distance from the barrier is still high.

TSUNAMI AT HILO, HAWAII

Hilo, Hawaii, is subject to severe damage from tsunamis which origi-

nate in the vicinity of Alaska and Chile. The orientation, topography
and hydrography of the region (Fig. 1*0 are such that it appeared likely
that a Mach-stem might have been associated with the 1 April 19^6 tsunami
which originated in the Aleutian Islands of Alaska. The height of the
tsunami should increase by a factor of k to 5 as it moves onto the shal-
low portion of the reef off Hilo, due to shoaling effects alone, and it
was believed that this, together with the Mach-stem effect, could account
for the characteristics of several of the waves of the tsunami as they
were observed. It was observed (M. L. Child of Hilo, Hawaii) that the
two waves that did the most damage came in as a bore in a southerly
direction along the cliff that forms the west border of the bay, swing-
ing easterly and running up through the streets of the town and into the
lee area of the breakwater, and waves also came over the top of the break-
water. A photograph taken at the time shows a wave that looks remarkably
 WATER WAVE EQUIVALENT OF MACH-RE FLECTION 93

2.4-

2.2

2.0

/
v
1.8
3*7*"
o <*

1.6
o

v/ A
•
1.4 A*.
x'
r-i» a
X* "O JL

15^ A+
1.2 -S^.**!.
/
cr;? sr D

/
1.0 £_

0.8

0.6 AngLe of incidence

10° 15° 20°

LEGEND:

Wave length Period Water depth Distance X along breakwater

L ft T sec d ft 2£ ft 54 ft
0.45
0.68
.30 .160 +
.38 .160 O
1.20 .58 .160 A A
1.46 .58 .240 • V
1.50 .71 .160 • •
FIGURE 9 INCREASE OP WAVE HEIGHT WITH ANGLE OP INCIDENCE.
Hj/d varies from 0.09 to 0.19
(from Nielsen, 1962)
WATER WAVE EQUIVALENT OF MACH-REFLECTION 95

(D) SEPARATED FREE WAVE

(a) SEPARATED WAVE

MACH STEM REFLECTION

( SHORT DURATION )

MACH STEM

MACH STEM
REFECTION

A)PURE
MACH STEM

FIGURE 12 -SCHEMATIC DEVELOPMENT OF WAVE

CRESTS ALONG VERTICAL CURVED BREAKWATER
WAVE HEIGHTS ARE INDICATED BY THICKNESS
OF CREST LINES ARROWS SHOW TRAVEL
DIRECTIONS (from Nielsen, 1962)
96 COASTAL ENGINEERING

in 0
TIVM A0y aN3 _a> CC
CO

o cc
0
ro -O HI
N i-
<
5 ^_
•x. 4-
C\J
< o u>
e UJ 0»
_o
to
a: — —
II
CD
T> c
sn
OJ
_|
<
V
in

o a>
0
*
_ Z
1-
-O
lO ce
UJ
d E
> ?i
$
a H
UJ X
O o >
cu -O
^" cc
3 _l
o UJ
Q
O
Z 5
°o q
_i o
~io
< z
in z
o
0
o I
o H o
*CM < cc
>
UJ
_l V)
0
UJ X
H-
"KT UJ <s>
O > z
< UJ
_l

UJ
2
3 3
S 5
X
< z
m s Ui
1 cc
UJ
ro u.
u.
Ul a
cc

o h u.
98 COASTAL ENGINEERING

like a Mach-stera (Fig. 13). Another photograph, showing the wave rolling
into tne town, is shown in Fig. 16.

From some of the observations of the tsunami of 22 May IStoO, origi-

nating in Chile, it appeared as if something similar must have happened,
out the autnor could not visualize how it could have occurred.

In order to study the gross characteristics of tsunamis at Hilo, a

1:15,000 (l: v'15,000 = 1:122 time and velocity scale) undistorted model
was constructed of fiberglass. The model was approximately 8 ft on a
siae, so tnat the entire bay could be included, as well as the reef. A
portion of the ocean was included, to a depth of 6,000 to 7>000 ft, pro-
totype. The model was placed at one end of an 8-ft wide by 6-ft deep by
20C-ft long tank so that a number of waves could be measured before
reflections could be of importance. A series of runs were made with
periods ranging from 6 to 24 minutes, prototype, and with waves from the
N, E, and SE directions.

It was found, for waves from the north, that a wave which had the
appearance of a Mach-stem was generated along the west cliff and rolled
into the town of Hilo in a manner that was similar to observations. It
was also found that the shoaling effect was about as theory predicted;
that is, the wave height increased by a factor of about k over the reef,
with respect to the wave height in the deep water portion of the tank.

After the tests had been run, it was brought to the investigator's
attention that due to refraction in the ocean, the tsunamis generated ofi
Chile would most likely approach Hilo Bay from an easterly direction,
rather than from a southeasterly direction as was originally supposed.
Because of this, the results of the model tests for the waves from the
east will be described herein. A remarkable phenomenon was observed in
the 12 to 20 minute (prototype) period range. Referring to Fig. lrfa,
the initial wave refracted to about the position shown as (l)-(l). The
northerly portion started to reflect from the coast while the southerly
portion continued to move towards shore. This resulted in the pattern
(2A)-(2A) as the reflected portion and (2)-(2) as the continuing portion.
As the reflected portion (2A)-(2A) moved down the coast, it became inde-
pendent of (2)-(2). At the same time the southerly tip of (2)-(2) dif-
fracted into the harbor, raising the water level. About at the same time
(2A)-(2A) progressed to position (3)-(3) with the portion near the coast
being considerably higher than the portion offshore. The portion near
the coast ran right along the coast, reaching positions (**-)-(4) and
(5)-(5) as a high wave running on top of the water which had diffracted
into the harbor from (2)-(2). It then ran into the town of Hilo. The
author believes that something similar to this must have happened during
the actual tsunami.

The transformation of (2A)-(2A) to (3)-(3) was probably caused by a

combination of refraction and reflection, together with some non-linear
effects because of the relatively large wave height to water depth ratio
along the coast, and the height of the tsunami at Hilo was probably due
to this combined with the diffracted wave. At some place between
 WATER WAVE EQUIVALENT OF MACH-REFLECTION 101

HILO
HILO

0 INCIDENT WAVE FRONT, REFLECTED WAVE, b MOTION OF MACH STEM

DIFFRACTED WAVE AND MACH STEM TRACED FROM ENLARGED 16mm
FILM, 1/8 SEC BETWEEN TRACINGS
(0 252 MINUTES, PROTOTYPE)

FIGURE 17, SOME RESULTS FROM A 1:15,000 SCALE MODEL STUDY OF

TSUNAMIS (16 MINUTE, PROTOTYPE) AT HILO HAWAII. RUN X, 8 FEBI963,
WAVE FROM THE EAST (IN 7,000 FT OF WATER, PROTOTYPE)
102 COASTAL ENGINEERING

(<2A)-(2A) ana (*>)-(i) '<=• i4aeh-st«:n type of pnenomencn evolved ana because
of its strength o came inaepenaent of the normally reflected portion of
the wave. In Fi^. 1 [b are shown the successive positions of this Mach-
stfam type of wave a , it moves along tne coast. Trierf. positions were
traced f'roTi enlargements of a 16 mm motion picture taken during the model
study, for a wave of b' second period (io minute period in the prototype).

ACKNOWLfc.DGMfc.NTo

The author wishes to acknowledge the valuable help of James D.

Cumming and Arne H. Nielsen for taking the data in the Hilo model tests,
and to W. J. Ferguson ana Arne H. Nielsen for constructing the model.

Rfc.FERfc.NCES

Chen, T. C, Experimental study on the solitary wave reflection along a

straight sloped wall at oblique angle of incidence, U. S. Army,
Corps of Engineers, Beach Erosion Board, Tech. Memo. No. 124, 24 pp.,
March 196l.

Friedlander, F. G., The diffraction of sound pulses, Proc. Roy. Soc.

(London), Ser. A, Vol. 186, No. 1006, pp. 322-367, 24 Sept. 1946.

Lighthill, M. J., The diffraction of blast. I, Proc. Roy. Soc. (London),

Ser. A, Vol. 198, No. 1055, PP- 454-470, 7 Sept. 1949-

Nielsen, Arne Hasle, Diffraction of periodic waves along a vertical

breakwater for small angles of incidence, Univ. of Calif., Berkeley,
IER Tech. Rept. HEL-1-2, 83 pp., Dec. 1962.

Perroud, Paul Henri, The solitary wave reflection along a straight ver-
tical wall at oblique incidence, Univ. of Calif., Berkeley, IER
Tech. Rept. 99-3, 93 pp., Sept. 1957.

Sigurdsson, Gunnar, and R. L. Wiegel, Solitary wave behavior at concave

barriers, The Port Engineer (Calcutta), pp. k-8, Oct. 1962.

Wiegel, R. L., Research related to tsunamis performed at the Hydraulic

Laboratory, Univ. of Calif., Berkeley, Proceedings of the Tsunami
Meetings associated with the Tenth Pacific Science Congress, Aug.-
Sept. 196l, Internat. Union of Geodesy and Geophysics, Monograph
No.- 2k, pp. 174-197, July 1963.

Wiegel, R. L., Oceanographical Engineering, Prentice-Hall, Inc., 535 PP«,

1964.